MicroRNAs are 17-24 nucleotide long single-stranded RNA molecules that profoundly impact the gene expression program of a cell by annealing to and suppressing the expression of many target genes. There are hundreds of microRNAs and different cell types express different levels of specific microRNAs. It has recently become apparent that changes in expression of microRNAs are involved in different types of malignancies. In addition, new families of short RNAs, like pi-RNAs, rasi-RNAs and tRFs are being discovered that are not related to microRNAs but also impact the phenotype of a cell. As part of the Mellon Urological Cancer Research Institute at the University of Virginia we are interested in studying and reversing prostate cancer progression. Prostate cancers are highly dependent on androgens for their proliferation. Androgens act on prostate epithelial cells by changing the gene expression program, but their effects on microRNA and other short RNA expression have not been studied. In addition, although prostate cancers are treated with and initially very responsive to anti-androgens, they usually recur as androgen-independent cancers that eventually metastasize and lead to death. To reverse this progression, we have to understand the mechanism by which prostate cancer cells become androgen-independent and acquire other features of progression. In this proposal we will clone and sequence by Solexa/Illumina libraries of small RNAs from prostate cancer cells to identify microRNAs and other short RNAs that are (a) regulated by androgens and (b) involved in prostate cancer progression. We will also follow the changes in known microRNAs by hybridization to locked-nucleic acid microarrays. These results will not only identify new short RNAs involved in prostate cancer progression, but also test the hypothesis that androgen-repressed short RNAs are often also repressed during progression of prostate cancer. We will then test the hypothesis that microRNAs repressed during prostate cancer progression contribute to the phenotype of progression, both in vitro and in vivo, starting with four microRNAs (in two families) that have already been validated in preliminary results. The final test of the hypothesis will be the identification of targets of these microRNAs whose de-repression during cancer progression leads to worsening of the malignant phenotype. We will identify the targets of two key microRNA families involved in cancer progression by a novel combination of combination of experiments and computational target prediction, starting with eight targets that have already passed the experimental filters. Thus new technologies, short RNA cloning and ultra-high throughput sequencing, and new combinations of assays for identification of relevant targets will be applied to the molecular analysis of prostate cancer progression and the function of key microRNAs in this process dissected. The results are expected to open new avenues of research, therapy and diagnosis of prostate cancers. PUBLIC HEALTH RELEVANCE: MicroRNAs and short RNAs that are repressed by androgens will open a new chapter in how androgens regulate gene expression. Specific microRNAs and short RNAs that are also repressed during prostate cancer progression can be increased for therapeutic purposes. Studying the effects of these microRNAs (and short RNAs) on target genes will reveal how progression of prostate cancer is accompanied by repression of microRNAs to de-repress specific target genes, rendering the cancer cells less dependent on androgens, more proliferative, migratory and invasive and more metastatic. Thus, these androgen-repressed and progression-repressed microRNAs and non-micro-short RNAs will be useful both as biomarkers of cancer progression and as tools to identify genes whose de-repression is critical for cancer progression. Two methods will be developed here that will improve the study of microRNAs in cancer: (1) The combination of cloning/sequencing with locked-nucleic acid microarrays to obtain a well validated list of microRNAs that are changed during cancer progression. (2) The combination of bioinformatics, microarrays of mRNAs and polyribosome fractionation of mRNAs to produce a list of targets with a high true-positive rate of being direct targets of the microRNAs.